Press Releases Archive   Search Press Releases
Press Release

F5 Networks' BIG-IP v9 is the Undisputed Performance Leader in Competitive Independent Third-Party Testing

BroadBand-Testing developed the industry's most comprehensive, diversified, and challenging test methodology to closely replicate real world scenarios and end-user experience; BIG-IP achieves Broad-Band's coveted Gold Award

F5 Networks, Inc. (NASDAQ: FFIV), the global leader in Application Traffic Management, today announced that BroadBand-Testing, a leading independent test lab specializing in the Layer 4-7 (L4-7) market, has deemed F5's BIG-IP® version 9 product the significant leader in its comprehensive comparative review of product performance. In addition to testing the BIG-IP product, the analysis included L4-7 networking products from Cisco, Nortel, Radware, NetScaler, and Redline. Based on the lab's rigorous analysis encompassing over 150 performance tests, BroadBand-Testing has awarded the BIG-IP product its esteemed Gold Award.

The test criteria established independently by BroadBand-Testing sets the highest industry standards for an objective analysis because it is based upon real-world scenarios, real clients, real servers, and real traffic. This testing-based on customer-driven requirements, challenges, and conditions-represents an industry-first due to its unique structure that allows virtually any interested customer to duplicate the testing and validate the results. Unlike traditional tests, BroadBand's test suite was designed to evaluate the real-world performance of a product in five key areas: L4, L7, compression, SSL, DDoS, and total system performance.

"This market was in real need of a comprehensive, fair, and transparent test methodology that is based on real world environments in order for customers to make sound buying decisions," said Joel Conover, principal analyst, Current Analysis. "The testing results clearly show the in-the-trenches performance that F5's BIG-IP can deliver to customers. The thoroughness, fairness, and scope of these tests and the fact that enterprises can duplicate them internally is a great win for the market."

Each test suite ran for over seven hours, and each device under test recorded several weeks of actual test time. This enabled BroadBand to evaluate the products in a real-world, day-in day-out usage pattern - a far more reliable assessment than simply running a one-off, five-minute test against a device. In addition each evaluated product was optimized for its highest possible performance. Using industry-leading traffic generation devices from Ixia and Spirent, F5's BIG-IP product out performed the competition to an astonishing degree.

The BroadBand report cites that the latency metric is absolutely critical, yet it is the most overlooked element (by vendors and customers) in performance testing. Vendors typically make maximum performance claims without mentioning the cost of that performance in terms of latency. For many users the latency performance hit is not worth the benefit of a high number of new sessions per second. Therefore, performance testing must measure such elements as Time to Last Byte (TTLB).

"It's common practice in our industry to create testing scenarios optimized to highlight a particular product's strength-that's not what happened here," said Jason Needham, Director of Product Management at F5 Networks. "By creating a truly level playing field, BroadBand-Testing took an entirely different approach. But not only were all products tested under the same conditions, those conditions were selected, configured, and proven to be the same situations faced by real customers in the real world. The fact that BIG-IP outperformed every other product when placed in these actual user scenarios is extremely gratifying and validation of the feedback we receive from customers every day."

In more than 150 individual performance tests conducted by BroadBand across five categories, the BIG-IP product came in first in all categories. Testing has proven that the BIG-IP product is unmatched in high throughput and low latency, even at high rates of traffic. This demonstrates that, while other solutions may have some similar functions as the BIG-IP product, when used in real world environments they are impractical because of the processing overhead they trigger. Highlights of the BIG-IP test results in each of the five categories are highlighted below:

Layer 4-BIG-IP delivers almost twice the speed of the nearest competitor. For both connections per second and throughput, BroadBand Testing saw a consistent set of results: the BIG-IP 6800 proved fastest in both tests. The implication here is that the BIG-IP product has less internal bottlenecks in its architecture, being able to perform well at all file size request points, and maintain that level of performance regardless of the metric being measured.

Layer 7- BIG-IP is the notable L7 leader with 110,000 connections per second. This metric shows that the product's performance holds up well moving from L4 to L7. This speaks well for using the BIG-IP product in a complex L7 environment with rules to contend with; meaning performance and intelligence can go hand in hand.

Compression-BIG-IP dominates throughput and response times. Throughput performance for the BIG-IP product was well over four times the nearest competitor and response times were up to 10 times better than the competition.

SSL-BIG-IP delivers two to 10 times SSL performance over competitors. The SSL tests resulted in record performance figures for the BIG-IP device, more than double anything that Broadband-Testing has seen, or what has been publicly published to date.

Distributed Denial of Service (DDoS)-BIG-IP is well ahead of the competition. A DDoS test was run using Spirent's WebAvalanche to generate the simulated DDoS attack, a SYN flood. The testing measured how the devices maintained performance (a L7 HTTP session with a 128 byte page request size) while under attack, with DDoS prevention functions enabled on all devices. The BIG-IP device emerges well ahead of the competition.

Total System Performance-BIG-IP overshadows all competitors in the 'combo' test. In this test scenario, all vendors' product features are set to 'ON' in a mixed traffic environment. In all cases the BIG-IP system came out significantly ahead of the group, with the nearest competitor achieving approximately 41% of the BIG-IP's total throughput and approximately 65% of BIG-IP's connection rate.

"Ixia's testing solution allowed BroadBand-Testing to create real-world performance testing scenarios that can be easily duplicated by anyone interested in replicating the results," said Sunil Kalidindi, Product Line Manager for Ixia. "We worked very closely with BroadBand to design the most realistic testing environment possible, using our IxLoad client and server emulation analysis solution for Layer 4-7 web stress testing. Executing more than 150 tests per device, running in excess of 7 hours. BroadBand truly pushed the envelope to make this test comprehensive and transparent."

About F5 Networks

F5 enables organizations to successfully deliver business-critical applications and gives them the greatest level of agility to stay ahead of growing business demands. As the pioneer and global leader in Application Traffic Management, F5 continues to lead the industry by driving more intelligence into the network to deliver advanced application agility. F5 products ensure the secure and optimized delivery of applications to any user - anywhere. Through its flexible and cohesive architecture, F5 delivers unmatched value by dramatically improving the way organizations serve their employees, customers and constituents, while lowering operational costs. Over 6,000 organizations and service providers worldwide trust F5 to keep their businesses running. The company is headquartered in Seattle, Washington with offices worldwide. For more information go to

# # #

This press release may contain forward looking statements relating to future events or future financial performance that involve risks and uncertainties. Such statements can be identified by terminology such as "may," "will," "should," "expects," "plans," "anticipates," "believes," "estimates," "predicts," "potential," or "continue," or the negative of such terms or comparable terms. These statements are only predictions and actual results could differ materially from those anticipated in these statements based upon a number of factors including those identified in the company's filings with the SEC.